A

i(1aGoao©

K

1

igialfDrc

2

0d

Journal of Ethnopharmacology 113 (2007) 125–131

Constituents, biological activities and quality controlparameters of the crude extract and essential oil from

Arracacia tolucensis var. multifida�

Mario Figueroa a, Isabel Rivero-Cruz a, Blanca Rivero-Cruz a,Robert Bye b, Andres Navarrete a, Rachel Mata a,∗

a Departamento de Farmacia, Facultad de Quımica, Universidad Nacional Autonoma de Mexico,Ciudad Universitaria, Mexico City, Coyoacan 04510, Mexico

b Instituto de Biologıa, Universidad Nacional Autonoma de Mexico, Ciudad Universitaria, Mexico City, Coyoacan 04510, Mexico

Received 14 December 2006; received in revised form 18 April 2007; accepted 11 May 2007Available online 18 May 2007

bstract

Bioassay guided fractionation of an antimycobacterial extract of Arracacia tolucensis var. multifida (Umbelliferae) led to the isolation ofsoimperatorin (1), osthol (2), suberosin (3), 8-methoxypsoralen (8-MOP) (4), herniarin (5), scoparone (6), umbelliferone (7), dihydroxypeucedanin8), 5-methoxypsoralen (5-MOP) (9), isoscopoletin (10) and scopoletin (11). The isolates were tested against Mycobacterium tuberculosis and only–4 showed significant activity with MIC values of 64, 32, 16 and 128 �g/mL, respectively. The essential oil showed moderate in vitro antibacterialctivity against representative Gram-positive and Gram-negative bacteria. The volatile oil of Arracacia tolucensis var. multifida was analyzed by

C–MS and found to be composed mainly by 2 and 3. The essential oil (IC50 = 116.4 ± 23.2 �g/mL) and the extract (IC50 = 1153.1 ± 53.2 �g/mL)f the plant provoked concentration dependent inhibition of the tone and amplitude of the guinea-pig ileum spontaneous contractions; the latterctivity was related with the high coumarin content of this species. A suitable (novel and rapid) HPLC method to quantify the major active coumarinsf the plant was developed. The method provides also a reproducible fingerprint useful for identity tests of this plant.

2007 Elsevier Ireland Ltd. All rights reserved.

bial a

assBffpsc

eywords: Arracacia tolucensis var. multifida; Spasmolytic activity; Antimicro

. Introduction

Arracacia tolucensis var. multifida Hemsley (S. Wats.) Math-as & Constance (Umbelliferae) is one of 38 species of theenus Arracacia found in the Sierra Madre mountains of Mex-co down through the Andes in South America. The plant is

tall perennial herb that grows commonly in the dry grass-ands and oak-juniper woodlands of Mexico ranging in altituderom 2250 to 2950 m over the sea level, from the States of

urango to Hidalgo south through Oaxaca. Local population

efer to this species by its Nahuatl name, “acocotli”, or theommon Spanish names of “comino rustico”, “hierba del oso”

� Taken in part from the MS and BS theses of M. Figueroa, UNAM 2005 and006.∗ Corresponding author. Tel.: +52 5 55 6225289; fax: +52 5 55 6225329.

E-mail address: rachel@servidor.unam.mx (R. Mata).

vnadfp

tt

378-8741/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved.oi:10.1016/j.jep.2007.05.015

ctivity; Coumarins; Antimycobacterial; Umbelliferae

nd “neldo”. The fruits and aerial parts of Arracacia tolucen-is var. multifida have been used as carminative and digestivetimulant agent along with Arracacia atropurpurea (Lehm.)enth. & Hook (Martınez, 1989). It has also been employed

or treating gonorrhea, fevers and anger (Argueta, 1994). Theruits have been employed in the past as a condiment and thelant eaten as food (Bois, 1904). Although Arracacia tolucen-is var. multifida has not been previously investigated from thehemical point of view, studies on the related species Arracaciaaginata and Arracacia nelsonii led to isolation of several pyra-ocoumarins, phenylpropanoids and monoterpenoids (Calderonnd Rıos, 1975; Delgado and Garduno, 1987). Recently weemonstrated that the crude extract of this species was not toxicor mice or mutagenic when tested by the Lorke and Ames

rocedures, respectively (Deciga et al., 2007).

Despite the continued use of Arracacia tolucensis var. mul-ifida its composition and pharmacological properties as well ashe quality control procedures for the crude drug have not been

1 noph

etpAipwpp

2

2

taoe(sioolcpHcwwd(rffia3otcofm7r(a

2

c(wo

BH

2

2

ttw9s

2

p1o

moFbfatccHa1

ecdaopC(r

eaCoot[p

cr

26 M. Figueroa et al. / Journal of Eth

stablished yet. Therefore, the present study was undertaken (i)o determine the potential antispasmodic action, chemical com-osition and active principles of the extract and essential oil ofrracacia tolucensis var. multifida and (ii) to develop an analyt-

cal method using HPLC to quantify the most important activerinciples of the plant. Altogether, the results of these studiesill be useful for establishing quality control and preclinicalharmacological parameters for the elaboration of scientific andharmacopoeic monographs of this Mexican medicinal plant.

. Materials and methods

.1. General experimental procedures

IR spectra were obtained on a Perkin-Elmer 599-B spec-rophotometer. 1H NMR (300 and 400 MHz) and 13C NMR (75nd 100 MHz) spectra were recorded in a Varian VXR-300Sr Varian Unit Inova spectrometer in CDCl3 using tetram-thylsilane (TMS) as an standard internal. EI mass spectraionization energy of 70 eV) were obtained on a HP 5890pectrometer. HPLC was carried out with a Waters HPLCnstrument equipped with Waters Dual 2487 detector. Controlf equipment, data acquisition, processing and managementf chromatographic information were performed by the Mil-enium 2000 software program (Waters). The analyses werearried out on a Nova-Pak® column (300 mm × 3.9 mm, 6 �marticle size, Waters). The mobile phase was an isocraticex–CH2Cl2–MeOH (80:7:13) system. The flow rate was kept

onstant at 0.6 mL/min for 30 min. All solvents and reagentsere analytical grade n-hexane, CH2Cl2, MeOH and EtOAcere obtained from Merck (Darmstadt, Germany) and Bur-ick & Jackson (Muskegon, USA). GHP® membrane filters0.45 �m) for the mobile phase were supplied by Pall Corpo-ation (New York, USA) and PVDF® membranes (0.45 �m)or the preparation of samples before HPLC injection wererom Whatman (Germany). GC–MS analysis was carried outn a JEOL JMS-AXOCCHA gas chromatograph interfaced to

Hewlett Packard 5890 mass spectrometer equipped with a0 m long × 0.32 mm i.d. × 0.30 �m film thickness composedf 5% phenylmethylsilicon HP column, connected to an ionrap detector operating in the electron impact mode at 70 eV;arrier gas was He, flow rate 1 mL/min and injection volumef 20 �L (in CH2Cl2). The oven temperature was programmedrom 150 to 300 ◦C with increase of 10 ◦C/min. Column chro-atography (CC) was carried out using silica gel 60 (Merck,

0–230 �m; ASTM, 0.063–0.200 nm). Thin layer chromatog-aphy (TLC) was performed on plates with silica gel 60 F254Merck, 0.25 mm); visualization of plates was carried out usingceric sulphate (10%) solution in H2SO4.

.2. Plant material

The aerial parts of Arracacia tolucensis var. multifida were

ollected southern Mexico City, Mexico on September 1999batch 1) and January 2004 (batches 2 and 3). The speciesas identified by Dr. Robert Bye from the Biology Institutef UNAM. Voucher specimens (Bye & Morales 27040 and

sf[0

armacology 113 (2007) 125–131

ye 33821, respectively) have been deposited at the Nationalerbarium of Mexico (MEXU), UNAM, Mexico City.

.3. Extraction and Isolation

.3.1. Essential oilThe essential oil (4 g) was obtained by hydrodistillation from

he aerial parts (batch 1, 1.0 kg). Major constituents of the essen-ial oil were identified by matching their 70 eV mass spectraith those of the reference library. In the case of 1, 2, 4 andthe identities were established by comparison with authentic

amples.

.3.2. Extraction and isolationThe air-dried aerial parts (batch 1, 2.0 kg) was ground into

owder and extracted by maceration with CH2Cl2–MeOH (1:1,0 L) at room temperature. After filtration, the extract was evap-rated under reduced pressure to yield 160 g of residue.

The organic extract was subjected to open column chro-atography on silica gel (1.0 kg) and eluted with a gradient

f Hex–EtOAc (1:0 → 0:1) and EtOAc–MeOH (1:0 → 0:1).ractions of 200 mL each were collected and then com-ined according their TLC patterns to yield seven primaryractions (F1–F7). Each primary fraction was tested for itsntimycobacterial activity. From fraction F1 (GI of Mycobac-erium tuberculosis = 99%) eluted with Hex–EtOAc (9:1)rystallized 500 mg (0.025%) of compound 1. HPLC purifi-ation of the mother liquors from fraction F1 [0.5 mL/min,ex–CH2Cl2–EtOAc (70:25:5)] yielded 2 (100 mg, 0.005%)

nd 3 (30 mg, 0.0015%). The retention times were 12.61 and9.50 min, respectively.

Fraction F2 (GI of Mycobacterium tuberculosis = 100%),luted with Hex–EtOAc (8:2), was subjected to open columnhromatography on silica gel (170 g) and eluted with a gra-ient of increasing polarity of CH2Cl2–MeOH (1:0 → 0:1),fforded five secondary fractions (F2-I–F2-V). Preparative TLCf F2-II using CH2Cl2–MeOH (9:1) yielded 6 (17 mg, 0.00085%,f 143–145 ◦C). HPLC purification of F2-III [0.35 mL/min,H2Cl2–MeOH (99:1)] afforded the known coumarins 1

50 mg, 0.0025%), 4 (70 mg, 0.0035%) and 5 (10 mg, 0.0005%);etention times: 18.76, 20.48 and 21.94 min, respectively.

Fraction F4 (GI of Mycobacterium tuberculosis = 96%)luted with Hex–EtOAc (1:1) was rechromatographed on

Si gel (300 g) open column eluting with a gradientH2Cl2–MeOH (1:0 → 0:1). Eight secondary fractions werebtained (F4-I–F4-VIII). Extensive TLC [CH2Cl2–MeOH (9:1)]f F4-IV yielded 7 (35 mg, 0.00175%, pf 230–232 ◦C). Fur-her purification of secondary fraction F4-V by preparative TLCCH2Cl2–MeOH (95:5)] led to isolation of 8 (25 mg, 0.00125%,f 134–135 ◦C).

Primary fraction F5 (GI of Mycobacterium tuber-ulosis = 99%) eluted with Hex–EtOAc (3:7) wasechromatographed on a Si gel column using a gradient

ystem of CH2Cl2–MeOH (1:0 → 0:1). Seven secondaryractions were obtained (F5-I–F5-VII). HPLC purificationHex–CH2Cl2–EtOAc (40:30:30)] of F5-II afforded 11 (12 mg,.0006%, Rt 9.49, pf 230–232 ◦C). Finally, preparative TLC of

noph

Fp

21

iocitsodtucQw(

2

St9fCraa1p57eacgi(igdaepiti

2

p

H(rTpebw

2

2

O1m1K1cwai(swswiwcweimswtcec(a

3

3

fMdt

M. Figueroa et al. / Journal of Eth

5-VII using CH2Cl2–MeOH (95:5) yielded 9 (20 mg, 0.001%,f 187–188 ◦C) and 10 (25 mg, 0.00125%, pf 178–180 ◦C).

.4. Quantitative analyses of the crude extract (batches–3)

Calibration curves were prepared independently by dissolv-ng the appropriate amount of each compound in CH2Cl2 tobtain final concentrations ranging from 0.75 to 25.0 �g/mL. Allompounds were detected at 285 nm. A volume of 20 �L wasnjected. The calibration curves were based on the peak areas ofhe HPLC chromatograms. The experiments were performed byextuplicate. The values were expressed in terms of percentagen the basis of the concentration injected. Furthermore, threeifferent concentrations of coumarins 1, 2, 4 and 9 were addedo a sample of the crude extract. The added concentrations weresed to confirm the amount of compounds 1, 2, 4 and 9 in therude extract by means of the standard addition method (ICH2(R1), 2005). Validation test materials (coumarins 4 and 9)ere purchased from Sigma (St. Louis, MO, USA) or isolated

1 and 2) as described above.

.5. Antimicrobial disk assay

Gram-positive bacteria [Bacillus subtilis (ATCC 6633) andtaphylococcus aureus (ATCC 25932)] and Gram-negative bac-eria [Escherichia coli (ATCC 10536), Salmonella typhi (ATCC992) and Pseudomonas aeruginosa (ATCC 27853)] were usedor antimicrobial tests. All microorganisms were provided byulture Collection at School of Chemistry, UNAM. The bacte-

ial strains were grown in Mueller-Hinton agar (MHA) platest 37 ◦C (Mitscher et al., 1987; NCCLS, 1997, 2003; Ojala etl., 2000; Alderman and Smith, 2001). Agar plates containingmL (106 bacteria/mL) of an overnight broth culture were pre-ared. Disks having a diameter of 6.0 mm were impregnated with�L of each sample at a final concentration of 100, 250, 500,50 or 1000 �g were placed on the inoculated plates. Similarly,ach plate carried a blank disk, with solvent only (DMSO) andntibiotic disk of amikacin (CIC = 0.15 �g/�L; Staphylococ-us aureus), vancomicin (CIC = 0.10 �g/�L; Bacillus subtilis),entamicin (CIC = 0.20 and 0.05 �g/�L; Pseudomonas aerug-nosa and Escherichia coli, respectively) and ciprofloxacinCIC = 0.015 �g/�L; Salmonella typhi). All the plates werencubated at 37 ◦C for 24 h. The susceptibility of each microor-anism to the samples (crude extract and essential oil) wasetermined by measuring the sizes of the inhibitory zones on thegar surface around the disks, and values <8 mm were consid-red as non-active against bacteria. All of the experiments wereerformed by cuadruplicate. The results are reported as criticalnhibitory concentration (CIC) defined as the lowest concentra-ion at which no bacteria growth inhibition was observed afterncubation at 37 ◦C for 24 h.

.6. Antimycobacterial activity

Activity of the crude extract, fractions and isolated com-ounds was determined against Mycobacterium tuberculosis

sata

armacology 113 (2007) 125–131 127

37Rv (ATCC 27294) in the Microplate Alamar Blue AssayMABA) as previously described. The growth of the bacteria isepresented as a numerical value called the growth index (GI).he percentage inhibition was defined as (1 − (GI of test sam-le/GI of control) × 100). The activity of pure compounds wasxpressed as the lowest drug concentration that affected an inhi-ition of ≥90% relative to untreated cultures (MIC). Rifampinas used as a positive control (MIC = 0.1 �g/mL).

.7. Spasmolytic activity

.7.1. Isolated guinea-pig ileum testAll experiments were performed following the Mexican

fficial Norm for Animal Care and Handing (NOM-062-ZOO-999). Each piece of ileum (1–1.5 cm long) from a freshly killedale guinea pig was suspended in a tissue chamber containing

0 mL of Tyrode solution with composition (mM): NaCl 149.2;Cl 2.7; MgCl2·6H2O 2.1; CaCl2 3.6; NaH2PO4 0.4; NaHCO31.9; glucose 5.0; pH 7.4, at 37 ◦C, continuously gassed witharbogen (95% O2 + 5% CO2). The contractions were recordedith an isometrical force transducer (Grass FT 03) connected toMP100 Manager Biopac System polygraph. The data were dig-

talized and analyzed by mean of software for data acquisitionAcknowledge 3.7.3). Tissues were placed under a resting ten-ion of 1.0 g and allowed to stabilize for 45 min and they wereashed with fresh Tyrode solution at 15 min intervals before

tarting the experiments. After the stabilization period the tissuesere contracted with carbachol (1 × 10−6 M) to assess its viabil-

ty. The extract and essential oil suspended in 0.5% Tween 80 inater were added to the bath in a volume of 100 �L at final con-

entrations of 3, 10, 30, 100, 300 and 1000 �g/mL. All responsesere recorded during 10 min. The antispasmodic activity of the

xtract was assessed by its ability to prevent the contractionsnduced by a submaximal concentration experimentally deter-

ined of acetylcholine (1 × 10−4 M), histamine (1 × 10−3 M),erotonine (1 × 10−4 M) and BaCl2 (1 × 10−2 M). Papaverineas used as positive control. All the results are expressed as

he mean of six experiments ± S.E.M. Concentration–responseurves (CRC’s) for the extract and oil were plotted and thexperimental data were adjusted by nonlinear least squares,urve fitting program (SigmaStat). The statistical significancep < 0.05) of differences between means was assessed by annalysis of variance (ANOVA) followed by a Turkey’s test.

. Results and discussion

.1. Antibacterial activity

Since Arracacia tolucensis var. multifida is used as an antiin-ective agent, a CH2Cl2–MeOH (1:1) extract was tested against

ycobacterium tuberculosis using the Alamar Blue proce-ure (Collins and Franzblau, 1997). The results revealed thathe extract significantly inhibited Mycobacterium tuberculo-

is growth (GI = 99%). Bioassay guided fractionation of thective extract using silica gel column chromatographic revealedhat primary chromatographic fractions F1–F5 concentrated thectivity (see Section 2.1). Extensive chromatographic separa-

128 M. Figueroa et al. / Journal of Ethnoph

t(i(ds1

tHtdmcltwcstmomwma

aiC

Tafbatf

mGdlatbtc9tta

3

aotwce(�2

3

tctetiidp(eatos

Fig. 1. Structures of coumarins 1–11.

ion of the active fractions led of the isolation of 11 coumarinsFig. 1) which were identified by IR, NMR and MS methods assoimperatorin (1), osthol (2), suberosin (3), 8-methoxypsoralen8-MOP) (4), herniarin (5), scoparone (6), umbelliferone (7),ihydroxypeucedanin (8), 5-methoxypsoralen (5-MOP) (9), iso-copoletin (10) and scopoletin (11) (Ivie, 1978; Harkar et al.,984; Estevez-Braun and Gonzalez, 2002).

In general the spectroscopic data were in agreement withhose previously described for compounds 1–11 (Ivie, 1978;arkar et al., 1984; Estevez-Braun and Gonzalez, 2002). Briefly,

he common feature of all structures is the benzo-�-pyrone unitiffering only in the nature of the substituents (i.e. hydroxy,ethoxy, isoprenyloxy, isoprenyl or furan) along the benzene

ore. The NMR data of compounds 1–11 showed strong simi-arities albeit partial. The 13C NMR and IR spectra confirmedhe �-pyrone unit since the characteristic carbonyl absorptionas observed at ∼160 ppm and ∼1720 cm−1, respectively. In all

ases, the 1H NMR spectra in the region between 6 and 8.2 ppmhowed the typical AB system for H-3 and H-4 and signals forhe aromatic. The nature of the substituents was determined by

eans of the analysis of the NMR spectra which revealed res-nances for methoxyl, hydroxyl, furan or different isoprenyloieties. The position of the substituents in the aromatic ringas established on the basis of the splitting pattern of the aro-atic hydrogens (Ivie, 1978; Harkar et al., 1984; Estevez-Braun

nd Gonzalez, 2002).

The isolates were tested against Mycobacterium tuberculosis

nd only compounds 1–4 showed significant anti-TBC activ-ty with MIC values of 64, 32, 16 and 128 �g/mL, respectively.ompounds 5–11 displayed MIC values higher than 128 �g/mL.

cc(a

armacology 113 (2007) 125–131

he better activity of 1–3 over the remaining compounds are ingreement with the observations of Schinkovitz et al. (2003) whoound that the presence of an isoprenyl unit attached to the car-ocyclic ring of some coumarins provoked an increment of theirntimycobacterial properties. Also these authors demonstratedhat umbelliferone (7) was weakly active against Mycobacteriumortuitum.

The extract and essential oil of Arracacia tolucensis var.ultifida was also tested against an appropriated battery ofram-positive and Gram-negative bacteria using the diskiffusion method. The crude extract was active against Staphy-ococcus aureus and Bacillus subtilis; the CIC values were 9.92nd 27.7 �g/�L, respectively. On the other hand, the essen-ial oil showed moderate in vitro antibacterial activity againstoth Gram-positive (Staphylococcus aureus and Bacillus sub-ilis) and Gram-negative bacteria (Salmonella typhi, Escherichiaoli and Pseudomonas aeruginosa) with CIC values of 50.4,9.8, 150.2, 150.1 and 147.9 �g/�L, respectively. The suscep-ibility of common Gram-negative and Gram-positive to theested material was poorer in comparison with that of standardntibiotics.

.2. Essential oil composition

The volatile oil of Arracacia tolucensis var. multifida wasnalyzed by GC–MS and found to be composed mainly bysthol (2), suberosin (3), 8-MOP (4), 5-MOP (9), benzyl alcohol,erpinen-4-ol and �-cadinene. Near a 50% of the constituentsere identified, being compounds 2 and 3 the major active prin-

iples. In the case of compounds 2–4 and 9 the identities werestablished by comparison with authentic samples. The Rt’sretention times) for compounds benzyl alcohol, terpinen-4-ol,-cadinene, 9, 4, 2 and 3, were 9.13, 11.66, 16.24, 23.27, 23.46,4.18 and 24.84 min, respectively.

.3. Spasmolytic activity

The use of Arracacia tolucensis var. multifida as carmina-ive and digestive stimulant agent suggested that this speciesould have spasmolytic properties. In order to corroboratehis hypothesis, the effect of both a CH2Cl2–MeOH (1:1)xtract and the essential oil prepared from the aerial parts ofhe plant on the spontaneous contraction of the guinea-pigleum was assessed. The results of the pharmacological test-ng revealed that both preparations provoked a concentrationependent inhibition of the tone and amplitude of the guinea-ig ileum spontaneous contractions (Fig. 2a). The essential oilIC50 = 116.4 ± 23.2 �g/mL) was more active than the crudextract (IC50 = 1153.1 ± 53.2 �g/mL), although three times lessctive than papaverine (IC50 = 35.0 ± 1.4 �g/mL) used as posi-ive control (Fig. 3). The influence of the extract and essential oiln histamine-, acetylcholine-, serotonine- and BaCl2-inducedmooth muscle contractions was also investigated (Fig. 2b and

). The crude extract as well as the essential oil antagonized theontractions exerted by acetylcholine (1 × 10−4 M), histamine1 × 10−3 M), serotonine (1 × 10−4 M) and BaCl2 (1 × 10−2 M)s is showed resumed in Table 1. Altogether this information

M. Figueroa et al. / Journal of Ethnopharmacology 113 (2007) 125–131 129

Fig. 2. Tracing of guinea-pig ileum muscle showing the response contractilebsa

rtfit

Fig. 3. Concentration–response curves showing the relaxatory effect of theextract and essential oil from Arracacia tolucensis var. multifida on the spon-tps

2ewsvtird

3

TEg

S

I

E

V

y (a) essential oil (1000 �g/mL), (b) spasmogen (BaCl2 1 × 10−2 M) and (c)pasmogen after treatment with essential oil (1000 �g/mL) during 10 min. Therrows indicate the treatment added to the bathing fluid.

evealed that the smooth muscle relaxant activity displayed byhe extract and essential oil of Arracacia tolucensis var. multi-da involved an unspecific inhibition of the calcium influx intohe smooth muscle cells (Rojas et al., 1995, 1996).

tv

able 1ffect of extract and essential oil from Arracacia tolucensis var. multifida on the contuinea-pig ileum

ample Acetylcholine (1 × 10−4 M) Histamine (1 × 1

C50 (�g/mL) ± S.E.M.Extract 38.7 ± 12.3 53.4 ± 32.5Essential oil 30.0 ± 2.8 23.7 ± 5.0Papaverinea 12.7 ± 4.1 12.8 ± 4.3

max (10−3 �g/mL) ± S.E.M.Extract 97.2* ± 1.1 98.2* ± 0.7Essential oil 98.8* ± 0.6 98.4* ± 0.4Papaverinea 100.0* ± 0.0 100.0* ± 0.0

alues are expressed as the percentages of inhibition of contractile responses calculaa Positive control.

aneous contractions of isolated guinea-pig ileum. Values are expressed as theercentages of inhibition of contractile responses calculated as the mean fromix different animals ± S.E.M., p < 0.05.

The presence of compounds 2 (Li et al., 1994; Chen et al.,000; Che-Ming et al., 2004), 4, 9 (Call and Green, 1956; Novakt al., 1967), 6 (Liu et al., 2002) and 10 (Reisch et al., 1966)ith known spasmolytic properties could account for the anti-

pasmodic activity of the extract and oil of Arracacia tolucensisar. multifida. Osthol (2) and 5-MOP (9) antagonized the con-ractions evoked by acetylcholine, histamine, BaCl2 and KCl onsolated rat ileum preparations. In addition, Teng et al. (1994)eported that osthol (2) inhibited calmodulin sensitive phospho-iesterase (PDE1) on trachea smooth muscle.

.4. Quantitative HPLC analysis

To develop a suitable HPLC fingerprint and a method to quan-ify the major coumarins of the extract of Arracacia tolucensisar. multifida, three different samples were collected in two dif-

ractions induced by acetylcholine, histamine, serotonine and BaCl2 on isolated

0−3 M) Serotonine (1 × 10−4 M) BaCl2 (1 × 10−2 M)

15.2 ± 5.1 27.8 ± 5.011.1 ± 1.5 15.4 ± 4.99.0 ± 1.8 11.3 ± 3.2

99.7* ± 0.1 99.3* ± 0.598.8* ± 0.1 98.9* ± 0.5

100.0* ± 0.0 100.0* ± 0.0

ted as the mean from six different animals ± S.E.M., *p < 0.05 (ANOVA).

130 M. Figueroa et al. / Journal of Ethnopharmacology 113 (2007) 125–131

Fig. 4. HPLC chromatrogam of crude extract from Arracacia to

Table 2Calibration data [regression equation and correlation coefficient (r2)] and LODfor coumarins 1, 2, 4 and 9

Analyte Regression equation r2 LOD (�g/mL)

1 y = 1.34 × 108x + 4.14 × 104 0.9999 0.472 y = 1.10 × 108x + 1.26 × 105 0.9999 0.1949

ftlrreto[bwaoftsb

TQ

C

1249

uo

4

pstgataccafscfi(d

y = 1.65 × 108x + 6.58 × 105 0.9989 0.15y = 1.53 × 108x − 2.45 × 105 0.9999 0.22

erent seasons (Fig. 4). The calibration curves (Table 2) indicatedhe linearity of the detector response for the four standard ana-ytes from 0.75 to 25.0 �g/mL. The limit of detection (LOD)anged from 0.15 to 0.47 �g/mL for the standard analytes. Theeproducibility and repeatability of the analytical method werevaluated in terms of the intermediate precision by analyzinghree replicates of six samples of stock solution (12.5 �g/�L)n 3 different days. The relative standard deviation (R.S.D.)n = 6] and coefficient of variation (CV) were calculated for eachatch. The results indicated that their chromatographic patternere similar showing in each case the presence of four peaks

ssigned as 1, 2, 9 and 4. The CV values (≤3.0) and the precisionf retention times and peak areas of compounds 1, 2, 4 and 9

or replicated injections. In order to determine the accuracy ofhe method, one sample was spiked with a known amount of thetandard analytes 1, 2, 4 and 9. Recovery ranges were found toe between 97.5 and 102.5%. Finally, this HPLC method was

able 3uantification of the coumarins 1, 2, 4 and 9 (mg/g of extract) by HPLC analysis

ompound Batch 1 Batch 2 Batch 3

9.8 ± 1.0 16.6 ± 0.8 15.8 ± 1.225.8 ± 4.0 60.2 ± 5.8 57.0 ± 6.467.4 ± 1.8 61.8 ± 3.8 58.8 ± 3.046.0 ± 5.8 42.0 ± 1.0 39.4 ± 1.4

tia

mpo

A

f

lucensis var. multifida of three different batches (b1–b3).

sed to quantify the coumarins 1, 2, 4 and 9 in the crude extractf Arracacia tolucensis var. multifida (Table 3).

. Conclusions

The organic extract and the essential oil from the aerialarts of Arracacia tolucensis var. multifida showed significantpasmolytic activity; this action could be useful to regulate gas-rointestinal disorders associated with indigestion or anomalousastrointestinal motility. The crude extract also showed notedntimycobacterial action against Mycobacterium tuberculosis;he latter effect could be related with the use of the plant as anntiinfective agent. The biological properties for Arracacia tolu-ensis found in this study were related with its high coumarinontent. Four of the coumarins isolated and characterized werective either as spasmolytic or antimycobacterial agents, there-ore, it is highly probably that these compounds contributedynergistically to the pharmacological actions exhibited by therude extract or essential oil of Arracacia tolucensis var. multi-da. The antimycobacterial activity of isoimperatorin (1), osthol2), suberosin (3) and 8-methoxypsoralen (8-MOP) (4) is newlyescribed.

The coumarins isolated from Arracacia tolucensis var. mul-ifida has not been isolated from the related species so farnvestigated which contain pyranocoumarins, phenylpropanoidsnd monoterpenoids.

A suitable (novel and rapid) HPLC method to quantify theajor active coumarins of the plant was developed. The method

rovides also a reproducible fingerprint useful for identity testsf this plant.

cknowledgements

This study was supported by the grants IN 212005rom Direccion General de Asuntos del Personal Academico

noph

(CDNiUts

R

A

A

B

C

C

C

C

C

D

D

E

H

V

I

L

L

M

M

N

N

N

O

R

R

R

S

M. Figueroa et al. / Journal of Eth

DGAPA-UNAM) and C01-018 from Consejo Nacional deiencia y Tecnologıa (CONACyT). We are indebted to G.uarte, R.I. del Villar and M. Gutierrez for recording mass,MR and IR spectra. The technical support of Dr. Jesus Arrieta

s acknowledged. Special thanks are due to Dr. Scott Franzblau,IC, Chicago, USA, who kindly arranged for the antimycobac-

erial evaluations. Mario Figueroa acknowledges a graduatetudent fellowship awarded by CONACyT.

eferences

lderman, D.J., Smith, P., 2001. Development of draft protocols of standardreference methods for antimicrobial agent susceptibility testing of bacteriaassociated with fish diseases. Aquaculture 196, 211–243.

rgueta, A., 1994. Atlas de las Plantas de la Medicina Tradicional Mexicana,vol. 1. Instituto Nacional Indigenista, Mexico City, Mexico, p. 340.

ois, D., 1904. Sur l’acocote (acocotli). Bulletin Societe Naturelle d’Acces 51,116–117.

alderon, J., Rıos, T., 1975. Isolation and identification of some constituents ofArracacia vaginata. Lloydia 38, 174–175.

all, T., Green, J., 1956. Spasmolytics from plants. I. Suksdorfin A andcolumbianin. Proceedings of the Montana Academy of Sciences 16, 49–51.

he-Ming, T., Cheing-Huang, L., Feng-Nien, K., Tian-Shung, W., Tur-Fu, H.,2004. The relaxant action of osthole isolated from Angelica pubescens inguinea-pig trachea. Archives of Pharmacology 349, 202–208.

hen, J., Chiou, W.F., Chen, C.C., Chen, C.F., 2000. Effect of the plant-extract[component] osthole on the relaxation of rabbit corpus cavernosum tissue invitro. Journal of Urology 163, 1975–1980.

ollins, L.A., Franzblau, S.G., 1997. Microplate alamar blue assay versusBACTEC 460 system for high-troughput screening compounds againstMycobacterium tuberculosis and Mycobacterium avium. AntimicrobialAgents and Chemotherapy 41, 1004–1009.

eciga, M., Rivero-Cruz, I., Arriaga-Alba, M., Castaneda, G., Angeles, G.,Navarrete, A., Mata, R., 2007. Acute toxicity and mutagenic activity ofmexican plants used in traditional medicine. Journal of Ethnopharmacology110, 334–342.

elgado, G., Garduno, J., 1987. Pyranocoumarins from Arracacia nelsonii.

Phytochemistry 26, 1139–1141.

stevez-Braun, A., Gonzalez, G.A., 2002. Coumarins. Natural Product Reports14, 465–475.

arkar, S., Razdan, T.K., Waight, E.S., 1984. Steroid, chromone and coumarinsfrom Angelica officinalis. Phytochemistry 23, 419–426.

T

armacology 113 (2007) 125–131 131

alidation of analytical procedures: text and methodology Q2(R1), TripartiteInternational Conference on Harmonisation Text (ICH), ICH Tech. Coordi-nation, Geneve, 2005.

vie, W.G., 1978. Linear furocoumarins (Psoralens) from the seed of Texas Ammimajus L. (Bishop’s Weed). Journal of Agricultural and Food Chemistry 26,1394–1402.

i, L., Zhuang, F., Zhao, G., Zhao, D., 1994. Effects of osthole on the contractivefunction of uterine smooth muscle. Xi’an Yike Daxue Xuebao 15, 164–167.

iu, H., Li, Z., Wang, X., Han, X., Teng, Z., Sun, Z., 2002. Effect of scoparone on[Ca2+]i of the isolated tracheal smooth muscle cells of guinea pig. ZhongguoYike Daxue Xuebao 31, 249–251.

artınez, M., 1989. Las Plantas Medicinales de Mexico. Editorial Botas, sixthed., Mexico City, Mexico.

itscher, L.A., Drake, S., Gollapudi, S.R., Okwute, S.K., 1987. A modernlook at folkloric use of antiinfective agents. Journal of Natural Products50, 1025–1040.

ational Committee for Clinical Laboratory Standards, 1997. Methods for Dilu-tion Antimicrobial Susceptibility Tests for Bacteria that Grows Aerobically,fourth ed., Approved standard M7-A4. National Committee for ClinicalLaboratory Standards, Wayne, Pa.

ational Committee for Clinical Laboratory Standards. 2003. Performance Stan-dards for Antimicrobial Disk Susceptibility Tests, eighth ed., Approvedstandard M2-A8. National Committee for Clinical Laboratory Standards,Wayne, Pa.

ovak, I., Buzas, G., Minker, E., Koltai, M., Szendrei, K., 1967. Isolation ofsome components from the herb of Ruta graveolens. Acta PharmaceuticaHungarica 37, 131–142.

jala, T., Remes, S., Haansuu, P., Vuorela, H., Hiltunen, R., Atela, K., Vuorela,P., 2000. Antimicrobial activity of some coumarin containing herbal plantsgrowing in Finland. Journal of Ethnopharmacology 73, 299–305.

eisch, J., Novak, I., Szendrei, K., Minker, E., 1966. Chemistry of natural sub-stances. V. Isoimperatorin, a component of Ruta graveolens. Pharmazie 21,628–629.

ojas, A., Cruz, S., Ponce, H., Mata, R., 1996. Smooth muscle relaxant com-pounds from Dodonaea viscosa. Planta Medica 62, 154–159.

ojas, A., Cruz, S., Rauch, V., Bye, R., Linares, E., Mata, R., 1995. Spasmolyticpotential of some plants used in traditional medicine for the treatment ofgastrointestinal disorders. Phytomedicine 2, 51–55.

chinkovitz, A., Gibbons, S., Stavri, M., Cocksedge, M.J., Bucar, F., 2003.

Ostruthin: an antimycobacterial coumarin from the roots of Peucedanumostruthium. Planta Medica 69, 369–371.

eng, C.M., Lin, C.H., Ko, F.N., Wu, T.S., Huang, T.F., 1994. The relaxant actionof osthole isolated from Angelica pubescens in guinea pig trachea. Archivesof Pharmacology 349, 202–208.